iee168fandomcom_de-20200215-history
Fabrikator User Tips: Andreas2015
1. Bed level Up to now, I have been working with the business cards, but using some feeler gauges seems a good idea. I am going to try that. 2. lower Z-axis end stop sensor This was the first bug, I found. As far as I read from the forums, every Fabrikator has it: The lower Z-axis end stop sensor can't be reached by the print bed, before the Z-axis thread nut hits the lower bearing. Setting the magnet on another position (there are two possible positions) doesn't help, because then the position of the upper Z-axis end stop sensor is wrong. My work around was to losen the lower z-axis sensor and mount it with the lower screw on the upper hole. For the sensor one srew is enough, as there is no mechanical load on it. ' 3. 5V voltage regulator' This issue nearly made me surrender with the Fabrikator. Only a few days after I started printing with the Fabrikator the electronic board quitted. (I described it here in the forum under "Fabrikator: power issue v3") What helped me was a hint from another Fabrikator user to check a microchip, called 5V voltage regulator (Thanks joop 1987!) I ordered some inexpensive voltage regulators via ebay and soldered it onto my board. (how-to videos are on youtube) It works fine, but the display has a little less contrast than before. However a brand new product shouldn't have such great problems. I was very disappointed. 4. inlet nozzle The main improvement I made to the Fabrikator, which I couldn't find elsewhere on the internet, was an inlet nozzle. When I solved the first issues and my printed parts became bigger, I often had the problem, that the filament bended out between the stepper motor and the hot end, disruppting the print immediatelly and instead conveying lots of filament onto my shopfloor. I wonder, why nobody else has this problem. Perhaps its a matter of the quality of the filament. However, I made a longer inlet nozzle from a M8-screw on my turning lathe, which I put in the hot end tube instead of the headless screw. It reaches almost to the brass wheel, preventing the filamant from bending. I might add a picture from my print head. What you can also see: I also marked the wheel with a black spot, to be able to see, when the wheel turns or not. Next you can see an alternative cooler holder (lower green part), only fixed with magnets for an easier nozzle cleaning and a sponge holder (also green) for swiping the filament. ' 5. step loss Y' Changing to bigger objects I also had the issue, that the Y-axis stepper motor lost steps, causing big offsets in the print, making the parts unuseable. It was always the Y-axis (as it is the heavier one), never the x-axis. And it occured when printing the filling of small walls, because then the printer makes a lot of criss-cross-movements and falls in a kind of resonance. At first I reduced acceleration and jerk-values on the Fabrikator software, which improved the situation, but not completely. Next I modified the stepper voltage after a hint from joop1987 from 0,35 to 0,4. Didn't have any loss since then. (Thanks again!) ' 6. step loss Z' I mention the step loss on Z separatelly, because it's completely different. During printing Z-loss didn't occure, because of the small moves. Z-loss only occured during the first 3 seconds (travel mode), when the table moves to its start height, but that is also important to work. So how to improve: First the thread nut: When assembling the printer according to the manual, you read a hint, that the thread nut shall look out from the bottom of the holder. Indeed there are 8 (!) different possible directions, in which the nut can be aligned. (Think about it!) For me it was the lowest one without a gap between spindle and nut. Second: Reduce the max Z-speed in software. It is not used during printing anyway, but only in travel mode at the start and at the end of printing. 7. ABS For ABS I found out the following parameters to work the best: nozzle temp: 230°C bed temp 110°C ' bed adhesion:' heated bed only: not sufficient, print slides from table, warpage makes the print lift from table. First I used Pritt stick, beeing able to run the printing I my living room. I worked fine, just be sure, that the pritt stick is applied just before printing and not the evening before. (It is still a little bit wet and sticky when starting printing, although the bed temperature is beyond 100°C. During printing the printed part really "bakes" on the glass plate and holds very strong.) I also reused the glue by brushing the dryed glue surface with water (outside from the machine), which always resulted in a big mess, but it worked. Next I made ABS juice from old ABS and acetone, It was the point, when my wife made me leave the living room and move to my workshop, as ABS juice is really aggresive and stinks! However for printing it works fine. During my first tries I read from "buildtak"-sheets, which you stick on your glass table. So I gave it a try and ordered one. They are not cheap, but work fine. The printing process becomes very easy, as you don't have to prepare anything before printing (unless pritt stick or abs juice) I wonder how long a sheet can be used, because despite of smaller scratchers from scrapers, I used to remove the print from the sheet, the buildtak sheet looks quite good. ' cooling:' With ABS this question is not so easy to clarify, for me and I think also for others. On smaller objects small details come clearer, when cooling is turned on. (There are test parts on Thingiverse, which allow to estimate the print quality) Also bridging over wider distances is no problem. Overhang up to 45° and more is possible. On greater objects (especially wide in X&Y) cooling leads to great warpage and the prints fall apart in slices. So cooling is turned off, with the negative aspect, that bridging looks terrible and almost no overhang is possible. Support structure is absolutely needed.) 8.PLA Sorry, no experience with PLA up to now. ' 9.rim' At first I needed it, I made 10 rims, like the default of Cura says. But with more experince it is not absolutely necessary, but only a waste of filament. I make 0 to 3, depending on the geometry. ' 10.skirt' should be used. Because on start there is often no filament in the nozzle. So the incomplete first round of printing becomes part of the skirt and not the print. And you can see the outline of your print. This make it easier to estimate, for example, if there is enough ABS juice on the table. ' 11.slicer' I prefere Cura, I like the 3D-preview, which slic3r doesn't have. It allows to recognize problems during slicing. Only when I want to print parts, that use the complete width of the table (150mmx150mm) I am forced to use slicer, as Cura only lets me print objects with 147x147. I haven't found the setting yet to improve that, also I use: no rim, no skirt, no raft